This application seeks funding to support the study of three HIV and drug prevention interventions with adult male inmates incarcerated within the Georgia Department of Corrections. Three specific aims will drive this research. They include: (1) examining the effects of the intervention components on the study variables, which will address intervention outcomes (i.e., sexual activities, drug use, and recidivism) as well as, health beliefs and behaviors (i.e., drug expectancies, self-efficacy for avoiding STD/HIV risks, future goal orientation, community and family involvement, and related psychosocial variables) believed to mediate these outcomes; (2) determining whether the hypothesized mediators explain, or statistically account for, differences in outcomes over time, and (3) assessing whether the effects of the interventions are moderated by characteristics of participants and/or processes associated with service delivery (e.g., number of sessions received, and client satisfaction). Studying this population has great public health significance since the majority of inmates return to communities and arming them with appropriate prevention information and skills can positively (we believe) impact the incidence and prevalence of HIV transmission and drug use. Little is known about the effectiveness of the interventions with the proposed population. Twelve cohorts comprising five hundred and seventy-six (576) soon-to-be-released inmates will be recruited to participate in the study with 144 assigned to each of the three intervention groups or a control group over the five year project. Inmates will receive a 12-session HIV and drug prevention intervention. Ex-offenders will be rigorously followed upon release at 3 and 6 month intervals to determine how they are adjusting to community living and to administer a battery of follow-up measures. The basic approach will employ a modified Latin-square research design. The data will be analyzed with inferential statistics (repeated measures ANOVA, MANCOVA, and covariance analysis to control for observed initial difference between intervention groups).